What’s Sharp?

Sometimes when I’m sitting out in our yard, I cover my right eye, to see what most people see. As it happens, my right eye has sharper than average visual acuity. With my glasses on, my left eye tests as 20/20, while the right is 20/10. That means ‘Ol Eagle Eye can discern a letter at twice the distance of my left eye – and the average person’s eyes.

The leaves high in the eucalyptus trees across the street are particularly interesting. With right eye covered, I can clearly tell what I’m looking at, with a sense of shape and color that make it easy to tell it’s eucalyptus leaves, but with no individual leaves clearly discernable as separate from the rest.

With the right eye uncovered, I can clearly see detail forms, edges and some central rib detail of individual leaves.

So, Moose, you may be thinking, other than braggadocio, what’s your point? Allow me to ease into the larger point with a small one. One of the things that drives me crazy about displaying images on the web is the difficulty of capturing the detailed texture of foliage. It tends to look mushy or over sharpened, to me, with little or no in-between.

I’ve posted about this on the Olympus list, without much interest from others. Perhaps because most people don’t see detail the same way I do.

If any of you with 20/20 vision look closely at an image that I’ve printed or prepared for the web, you might easily conclude that I’ve over sharpened it. And I may have – for your eyes, but perhaps not for mine.

Let’s assume that my friend, Bob N. Vision, and I stand looking at a lovely landscape. We are looking at the same thing, but what we are each seeing is somewhat different. It’s so pretty that we each take a pic with our SuperRez D9900 DLSRs.  After we recuperate from our trek with a couple of beers, Bob manages to pop his CF card out and lose it down a sewer grate. Well, no big deal, we took pretty much the same pics. So I send him copies of my RAW files when I get home.

A couple of days later, we both post that wonderful landscape picture in our blogs. Then Bob drops me an email suggesting ever so politely that mine is over sharpened. Being even more polite than he is, I haven’t even mentioned to him how soft his version looks to me.

We agree that web images aren’t that great anyway and arrange to bring prints to our next meeting. But guess, what? We each have the same opinions of each other’s prints.

So who’s right? Both of us I suppose. We have each created a print that tries to replicate how the subject looked to us – and indeed how all such subjects appear to each of us.

What Do We See When We See?

I’m talking literally here, not about some artistic issue about whether one “sees” or just looks. That’s interesting stuff, but I’m down here at a more basic level at the moment. Although in the broadest sense like a camera, with optical system that focuses an image on a sensor, the human eye is also quite different in detail.

Although the retina (sensor) is composed of individual sites that react to light individually, there are at least four different kinds, they are not distributed evenly and there is a largish spot not far off center with no sensor sites at all. One may assume that the eyes cover for each other. I suppose that’s true when both are working. But cover one eye and look for the blank spot. Not there unless you use one of the simple tests, which are sort of creepy.

The part of the retina directly behind the center of the lens in densely packed with small, low sensitivity sensors, cones, which also sense color. This area of high resolution, full color sensors has a visual field of view of about 2°, about 1% of the full horizontal visual field of a person with good peripheral vision. Outside of this central zone, the cones start to be mixed with larger, more sensitive, monochromatic sensors, rods. The further from the center, the more the mix becomes dominated by rods until all the outer areas are strictly low resolution and sense brightness only, no color.

Now stop for a moment, leave your eyes looking at a point in this text, and pay attention to your visual field. If your vision is like mine, a handful of words are clear at once, but to see a whole line, I have to move my eyes. This motion is usually unconscious, but clear if I pay attention.

Outside that sharp, clear area is a broader area where the sharpness and size of individual words is clear, with progressively less detail further from the center of focus. Even further out, there is a clear sense of seeing what’s there and of color, but when I force my eyes to stay in one place, the detail of this large peripheral area is limited, and I’m not sure about color. It seems like I’m seeing color, especially in bright objects, but I’m not sure.

Now another little detail. When focusing closely on something, our eyes make continuous micro-movements, of which we are unaware. This is necessary for the chemical process of light sensing to work continuously.

So, although we have a sense of a full, 3D, detailed, colorful visual field, it appears that it is in large part virtual image. At any given moment, only a small part of the incoming data is sharp and in full color, data is missing entirely for a significant part of the field covered by each eye. And when paying close attention to one thing, the image is jittering across the retina, so the data stream is not continuous in the sense of one we could get from a camera sensor.

The Visual Cortex - Post Processing Wetware

The conscious visual field we experience is not something that has ever been captured, measured, whatever. It is entirely inaccessible to anything or anyone other than each individual consciousness – something else that is also not understood or explained by science.  There are, of course, books on the subject. But so far, they are just speculation and theory that is not testable.

Another interesting thing about vision is that even internal visualization of a visual field unrelated to actual experience and access to visual memory depend on a functioning visual cortex. If sight is lost, that area starts, in as little as a few days, to switch function to other tasks. Eventually, visual memory disappears.

So this apparently solid vision we have and depend on is really rather ephemeral. Perhaps more interesting, it depends entirely on what many would call post capture processing. We can no more “see” what comes to our brain from our eyes without processing than we can visualize the content of a RAW file from staring at the bits in the file. Both are latent until processed.

So What’s Your Point?

Oops, darn, I know I had one around here somewhere. Oh, there’re a couple now.

- The unstated assumption that individual human vision systems differ less than they actually do may lead to misunderstandings over what seem to be differences of accuracy, intention and taste, which are in fact physiological.

- Superficial similarities aside, a camera is nothing like the human vision system. Any idea that the image as captured by a camera is somehow purer than one adjusted to reflect the way the photographer saw the subject, and possibly the image being created, is logically suspect.

This is only a start on those aspects of the human vision system and other aspects of the human condition that influence what and how we see. There are other, broader aspects and possible conclusions about seeing and photography to explore.

Moose de Bezerkeley
© Moose, 2008. All rights reserved.